Viscosity: A lubricant’s most important characteristic
Introduction
For any oil lubrication system, oil viscosity is considered as the most important parameter. One
should always ensure that the viscosity of the oil in use meets OEM recommendations.
The main function of a lubrication oil is to create and maintain a lubrication film between two
moving metal surfaces and this function is very much dependant on the viscosity of the oil itself.
Figure 1 illustrates the gap between two metal surfaces moving in opposite directions, separated by
a layer of lubricant.
Figure 1: Lubrication film for two moving surfaces.
Often, when the oil viscosity is not within specified viscosity range, a condition known as insufficient
lubrication will occur, resulting in increased friction, wear and heat. Figure 2 and 3 shows the result
of insufficient lubrication on inner race and rollers of a bearing, resulting in catastrophic failure.
Figure 2: Damaged inner ring of a
spherical roller bearing caused by
insufficient lubricant.
Figure 3: Damaged bearing rollers
caused by insufficient lubricant.
Failures such as this may be prevented if abnormal viscosity is detected early and the fault is
corrected.
Viscosity is one of the compulsory tests for routine in-service oil analysis. Any significant change
detected in the measured viscosity requires prompt action and could be indicative of severe
degradation of oil, cross contamination, water ingression or other factors that can be confirmed by
measuring other parameters (i.e. water content).
What is Viscosity
Viscosity can be defined as measurement of fluid internal resistance to flow at a specified
temperature. There are two ways to measure a fluid’s viscosity, namely Dynamic (Absolute) Viscosity
and Kinematic Viscosity.
Dynamic (Absolute) Viscosity
Dynamic Viscosity is defined as a fluid’s resistance to flow, or the fluid’s resistance to deform when
subjected to a force1. An easy way to visualise this is to imagine stirring two liquids, water and
honey, in two separate containers. The honey provides more resistance to shear forces through the
stirring process compared to water, and is said to have higher dynamic viscosity compared to water.
Dynamic viscosity is commonly reported in Centipoise (cP). 1 cP = mPa.s
Brookfield rotary method ASTM D2893 is the most common method to measure dynamic viscosity
as shown in Figure 4.
Figure 4: Brookfield rotary method ASTM D2893.
Kinematic Viscosity
Kinematic viscosity, defined as a fluid’s resistance to flow, is traditionally measured by noting the
time taken for a fluid sample to travel through an orifice in a capillary under the force of gravity
(Figure 5)2. This method is defined by ASTM D445 and currently used by most laboratories
worldwide. The time taken is noted and converted into Kinematic Viscosity, reported in Centistoke
units (cSt). 1 cSt = 1 mm2/s.
Figure 5: Method for Kinematic Viscosity
Figure 6: Automatic Kinematic Viscometer
It is important to note that most laboratories report viscosity as Kinematic Viscosity, whilst most onsite equipment reports in Dynamic Viscosity units. However, both Dynamic and Kinematic Viscosity
are interchangeable by using the formula below:
Dynamic Viscosity (cP) = Kinematic Viscosity (cSt) x Fluid Density (kg/m3)
Oil Viscosity Grade
Lubricants are classified according to their viscosity at 40°C by authorised bodies such as the
International Organisation for Standardisation (ISO), American Gear Manufacturer Association
(AGMA), Society of Automotive Engineers (SAE) etc. Figure 7 shows the classification of oil based on
viscosity.
Figure 7
It is important to note that the viscosity of a lubricant will decrease as the temperature increases.
Viscosity Index (VI) is an arbitrary scalar value that indicates the change in an oil’s viscosity with
changes in its temperature. A high Viscosity Index indicates less viscosity change when temperature
increases, indicating better resistance to thinning, for a given oil and, likely better film strength
retention under heat duress.
Figure 8: Comparison between viscosity change of two oils against temperature
Based on Figure 8, Oil A has higher Viscosity Index compared to Oil B.
Can we measure oil viscosity on-site?
There are several methods to measure viscosity on-site.
1. Viscosity comparator a.k.a. viscostick
The low cost, entry level of viscosity measurement. New and in-service oil is put into the viscostick
and tilted to allow both oils to flow down the channels. By comparing the distance that the two oils
flow, the result is either more viscous or less viscous than the reference oil. However, the result is
not specified in quantitative units. If the in-service oil viscosity is outside acceptable range, typically
a 10% difference to the reference oil, this is indicative of a change in viscosity. For more subtle
differences, a quantitative measurement would be needed.
Figure 9: Viscostick
2. Viscotube
Figure 10: Viscotube
A Viscotube uses “falling ball” technique to measure the viscosity of oil. Kittiwake’s DIGI Viscotube
uses this method to measure oil viscosity between 20-6000 cSt @ 40°C. The time taken by the ball to
fall inside a tube filled with an oil sample due to gravity force is noted. The data is inputted into the
supplied viscosity calculation software together with oil temperature, Viscosity Index and density.
The Viscosity calculation software automatically calculates the kinematic viscosity of an oil sample at
room temperature, 40°C, 50°C and 100°C (Figure 11).
Figure 11: Kittiwake viscosity calculation software
3. Viscometer
Figure 12: Kittiwake Heated Viscometer
Figure 13: Kittiwake Unheated Viscometer
Viscometers designed and produced by Parker Kittiwake are currently among the most accurate onsite instruments to measure viscosity and correlate strongly to laboratory measurements. Two
versions are available, heated and unheated.
With the heated version, the oil sample inside the viscometer is heated to 40°C. After the viscometer
is tilted, the ball inside the viscometer will drop due to gravitational force. The time taken by the ball
will be detected and calculated into oil kinematic viscosity in cSt automatically.
For the unheated version, the ball travel time is measured at room temperature, calculated and
corrected to kinematic viscosity at 40° automatically.
Both Heated and Unheated viscometers can calculate the kinematic viscosity at 100°C by inserting
the oil Viscosity Index value. For engine oil, both devices are able to display the oil viscosity
according to its SAE Grade.
The Viscosity range for the unheated version is between 20 and 810 cSt at 40°C and for the heated
version, between 20 and 810 cSt at 40°C.
Why oil viscosity monitoring is very important?
Most components surfaces are separated by a lubrication film thickness of 10 micron or less.
Especially for high pressure systems, such as hydraulics or high speed turbochargers, to maintain the
correct oil viscosity is even more critical. Any significant change of oil viscosity, either a reduction or
increment, may damage lubrication film stability and effectiveness. The results can be summarised
as Table 1:
Reduction in Viscosity
Loss of oil film resulting in excessive wear
Increased mechanical friction causing excessive
energy consumption and heat generation
Increased sensitivity to particle contamination due to
reduced oil film thickness
Oil film failure at high temperatures, high loads during
start-ups or coast-downs
Increase in Viscosity
Excessive heat generation resulting in oil oxidation,
sludge and varnish build-up
Gaseous cavitation due to inadequate oil flow to
pumps and bearings
Lubrication starvation due to inadequate oil flow
Oil whip in journal bearings
Excess energy consumption to overcome fluid
friction
Poor air detrainment or demulsibility
Poor cold start pumpability.
Table 1: Effect of significant oil viscosity increase and reduction 3.
The end results of above effects are shorter oil lifespan, shorter components lifecycle, increased oil
consumption, higher power consumption and reduced machine reliability.
It is extremely important for the end user to find the root cause of oil viscosity change and to act
accordingly at the earliest possible time to minimise the damage. Root causes are listed in Table 2:
Reduction in Viscosity
Increase in Viscosity
Thermal cracking of oil molecules
Oxidation
Shear thinning of VI improvers (for multigrade
engine oil)
Water (emulsion)
Formation of carbon and oxides insoluble
Fuel dilution
Soot
Cross mixing with lower viscosity oil
Antifreeze (Glycol)
Cross mixing with higher viscosity oil
Table 2: Root cause of viscosity change
Most of the factors, except water ingression, may be rectified by changing the oil. However, further
analysis is required to confirm the root cause before any actions are taken.
Once the root cause(s) is / are confirmed, remedial action should be taken to prevent the incident
from recurring, especially if contamination is identified.
It is important to note that sometimes two factors may be acting in opposite directions, resulting in
an acceptable viscosity range, for example soot (increases viscosity) and fuel dilution (reduces
viscosity) for engine oil. Both factors can be measured separately and if either of them exceed
acceptable limits, this may be detrimental to the machine. Hence, viscosity analysis alone is not
comprehensive.
How to set the viscosity limit for in-service oil
The first step is to set a baseline value as a reference. This value is based on actual viscosity
measurement of new fresh oil and not from the lubricant product sheet. Industrial oil (hydraulic,
turbine, compressor, gear etc) should be based on kinematic viscosity @ 40°C, whilst for engine oil,
kinematic viscosity @ 100°C should be used. The upper limits and lower limits for in-service oil
should be based on this value.
Noria recommendations set the limits (in percentages) as per Table 3.
Limit
Crankcase Oils**
Industrial Oils**
Severe Environment Industrial Oils**
Critical (upper)
+20%
+10%
+7%
Caution (upper)
+10%
+5%
+4%
Caution (lower)
-5%
-5%*
-5%*
Critical (lower)
-10%
-10%*
-10%*
* Twice this amount for oils with VI improvers.
**Crankcase oil limits based on cSt @ 100°C, industrial oils based on cSt @ 40°C.
Note: Severe environment oils are at a higher risk of thermal and oxidative degradation.
Table 3: Limitation for in-service oil viscosity 4.
Once the baseline and limits have been set, routine viscosity can be analysed by trending. Figure 14
shows one example of viscosity trend. It is important to note that small fluctuation in the measured
value is normal, for example after every top off activity. However, any significant change above the
limits requires an investigation.
Figure 14: Example of viscosity trend analysis
Figure 15 shows a chart of how viscosity monitoring should be performed, as well as the potential
investigation required.
Figure 15: Viscosity analysis and further action 4.
Case Study
Gearbox for Ship Loading Conveyor
Prior to December 2003, the gearbox was filled with ISO VG 150 Gear Oil, against the OEM
recommendation of ISO VG 680 Gear Oil.
Due to high wear rates and a laboratory recommendation, the gear oil was changed to the correct
grade ISO VG 680 in December 2003. However, this resulted in an increased wear rate, as per figure
16. High amounts of iron and lead indicate a bearing problem. This result was confirmed using
vibration testing.
Figure 16: Relationship between oil viscosity and wear metals amount in oil sample
A new gearbox was ordered and the running gearbox experienced a catastrophic failure one day
after the new gearbox arrived, exactly two weeks after the order was made.
A teardown investigation of the gearbox found that the channels designed as the lubricant’s path to
the bearings were plugged by sludge. Due to the lower viscosity oil that had been used, the ISO VG
150 lubricant was able to work past the sludge to lubricate the bearings just enough to forestall
excessive wear. When the lubricant was changed to the manufacturer-recommended ISO 680
lubricant, the sludge was sufficient to block the flow of lube to the bearings. This quickly resulted in
the catastrophic wear conditions seen 5.
The lesson learnt is by simply changing oil viscosity to OEM recommended grade does not guarantee
a reduction in excessive wear as the result. In this case, the root cause was blocking of the lubricant
channel by sludge, contributing to lubricant starvation and excessive wear that led to catastrophic
bearing failure. The proactive solution in this case is to maintain the correct oil viscosity at all times
and ensure no restriction of oil flow to the bearing.
Conclusion
As the most important characteristic for lubricating oil, viscosity of in-service oil must be monitored
as a routine parameter in oil analysis. However, in order to get the most benefits from oil analysis,
other important parameters must be analysed as well to get the holistic view of oil and machines
condition.
References
1. 2 ways to measure oil viscosity by Ashley Meyer. Practising Oil Analysis, (9/2007).
2. Understanding Absolute and Kinematic Viscosity by Drew Troyer. Practising Oil Analysis,
(3/2002).
3. Oil viscosity – How it is measured and reported by Noria. Practising Oil Analysis, (11/2002).
4. Trouble-shooting Viscosity Excursion by Jim Fitch. Practising Oil Analysis, (5/2001).
5. Gearbox oil analysis study by Matt Spurlock. Practising Oil Analysis (11/2005).
* Azhar bin Abdullah – Field Engineer, Kittiwake Asia Pacific Sdn. Bhd. Azhar has worked in the field
of Oil Condition Monitoring for 8 years. He currently works as a field engineer for Kittiwake,
demonstrating to and training customers in the use of oil condition monitoring equipment and
techniques. He holds a BSc in Mechanical Engineering (Universiti Teknologi Malaysia) and is a
certified ICML MLA Level III Oil Analyst.
* Steve Dye is the business development manager for Parker Kittiwake, concentrating on promoting
high end products into the market, including FTIR analysers. He has worked in high technology
businesses for all of his career including Hewlett Packard and Lucent Technologies, covering roles
from development and product management through to sales and marketing director. He has been
involved in the field of Condition monitoring for the past 3 years. Steve holds a Bachelors degree in
Communications Engineering and a PhD in Optoelectronics. He is also a ICML Level I certified
Lubrication Analyst.
* Jack Poley – Technical Director, Kittiwake Americas. Jack has over 50 years in Oil Analysis and is
recognised as a world expert in both laboratory and field measurement techniques. A member of
ASTM and STLE for over 35 years, Jack co-founded STLE’s Condition Monitoring Education Course. He
also co-founded the OMA (Oil Monitoring Analyst) certification program at STLE and is certified OMA
1 and OMA 2. Jack holds a B.S. in Chemistry (University of California [Berkeley]) and a B.S. in
Management (New York University School of Commerce).